Open Access
June, 1946 An Experimental Design for Slope-Ratio Assays
C. I. Bliss
Ann. Math. Statist. 17(2): 232-237 (June, 1946). DOI: 10.1214/aoms/1177730984


When the response to a drug is a linear function of arithmetic dosage units, the relative potency of two preparations can be computed as a slope-ratio assay. Their dosage-response curves are computed by solving three simultaneous equations to obtain the common intercept $a'$, the slope of the standard, $b_1$, and the slope of the unknown, $b_2$. The method is applicable to certain microbiological assays for the vitamins. Usually several unknowns are assayed at one time with a single standard. Their calculation is simplified when such assays meet the following requirements: (1) restriction of treatments to the zone within which the response is related linearly to the dose, (2) equal spacing of doses on an arithmetic scale beginning with the negative control, (3) an equal number $(k)$ of doses of standard and of each unknown and (4) $r$ replicates for each dose of unknown, $h'$ replicates for the negative control and $h$ replicates for each dose of the standard.


Download Citation

C. I. Bliss. "An Experimental Design for Slope-Ratio Assays." Ann. Math. Statist. 17 (2) 232 - 237, June, 1946.


Published: June, 1946
First available in Project Euclid: 28 April 2007

zbMATH: 0063.00458
MathSciNet: MR16627
Digital Object Identifier: 10.1214/aoms/1177730984

Rights: Copyright © 1946 Institute of Mathematical Statistics

Vol.17 • No. 2 • June, 1946
Back to Top